arXiv:2610.13784
Locality-Preparation Time Tradeoffs in Dissipative Quantum Systems
Philipp Westhoff, Sara Murciano, Mattia Moroder
Abstract
Dissipative state preparation engineers quantum many-body states as steady states of an open-system dynamics. The spatial range of the required dissipation sets both the feasibility of analog implementations and the connectivity and control complexity of digital ones. Yet how this locality quantitatively constrains preparation times remains largely unknown. Here, we establish an exact locality-preparation time relation for quadratic fermionic systems and develop a construction that optimizes the locality of the dissipative dynamics while preserving the target state. Applied to the Kitaev chain, our framework reveals sharply different locality-preparation time tradeoffs in critical and gapped regimes. At criticality, suppressing the algebraic tails of the jump operators necessarily forces the preparation time to grow algebraically with system size. By contrast, both trivial and topological gapped phases admit exponentially localized dissipation with a finite preparation time in the thermodynamic limit. Our results identify when fast dissipative preparation is compatible with local control and when criticality imposes an unavoidable dynamical cost.
AI Summary
The paper asks how the spatial locality of engineered dissipation constrains the time required to prepare fermionic ground states. It studies quadratic Hamiltonians with linear Lindblad jump operators formed by mixing Bogoliubov annihilation operators. Full-rank mixing preserves the target quasiparticle vacuum; the mixing is then chosen to improve real-space localization. For translation-invariant periodic systems, the mixing is parametrized by momentum weights $\beta_k$, normalized by $\max_k|\beta_k|=1$. The dissipative gap is $\Delta_{\mathcal L}=(\kappa/2)\min_k|\beta_k|^2$. Fermion-parity superselection excludes the slowest odd modes, giving a physical asymptotic relaxation timescale $\tau=[\kappa\min_k|\beta_k|^2]^{-1}$. Fourier regularity connects these rates to jump-operator tails: at the Kitaev chain’s Ising and XX critical lines, suppressing tails to $r^{-(n+1)}$ requires order-$n$ zeros in the momentum weights, yielding $\tau\sim N^{2n}/\kappa$. Both trivial and topological gapped phases instead admit exponentially localized jumps with $\tau=\mathcal O(1)$. On $\Delta^2+(\mu/2)^2=t^2$, an explicit construction has two-site support. The results follow from quadratic-Lindbladian spectral analysis and the analytic structure of Bogoliubov coefficients. Finite-size calculations also examine truncating jumps to finite range: smoothing before truncation can improve critical-state correlations and overlap, but excessive suppression can amplify truncation errors, producing nonmonotonic accuracy. The analysis concerns quadratic, Markovian dynamics, with explicit tradeoffs developed for translation-invariant chains. Its preparation-time measure is an asymptotic relaxation scale rather than a fixed-error global mixing time. Interactions, measurements and feedback, non-Markovian reservoirs, and implementation constraints remain open extensions.
Central claim may not hold: The advertised general, exact tradeoff is supported in the supplied derivation only within a restricted translation-invariant linear-jump construction, without a general optimality theorem. The claimed finite preparation time in gapped phases is a finite inverse relaxation rate, whereas preparation to fixed global accuracy can require logarithmically increasing time even for independently damped modes. At Ising criticality on an even-size periodic momentum grid, the proposed suppression weight vanishes exactly at k=pi, leaving an undamped mode rather than the claimed positive relaxation rate proportional to N^(-2n). The displayed finite-range jumps on the special ellipse omit the overall normalization needed to maintain the paper's fixed mixing-matrix norm and rate convention.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 6 pages, 3 figures in main text; 14 pages, 4 figures in supplemental material